Automated Medical Literature Summarization Using Workflow Automation and Large Language Models: A Proof-of-concept Study

Background: The exponential growth of biomedical literature creates significant challenges for clinicians and researchers.While large language models (LLMs) demonstrate remarkable text processing capabilities, their integration into practical clinical workflows remains limited.This proof-of-concept study describes the development and pilot evaluation of an automated literature processing system combining workflow automation with LLM technology.Materials and methods: We developed a proof-of-concept study integrating n8n workflow automation platform with OpenAI's GPT-4o-mini API.The system comprises four sequential stages: PDF content extraction, paragraph-wise summarization, content simplification to a medical student comprehension level, and structured review article generation formatted to ICMJE standards.The workflow was pilot-tested with sample medical literature to demonstrate technical feasibility.Results: We successfully developed and deployed a functional automated literature processing pipeline.Pilot testing demonstrated successful end-to-end workflow execution including PDF text extraction, multistage LLM processing, and automated e-mail delivery.Qualitative review of generated outputs confirmed preservation of clinical terminology, logical section organization, and ICMJE-compliant formatting.The study demonstrated technical feasibility for automating literature summarization workflows.Conclusion: This proof-of-concept study demonstrates the technical feasibility of integrating workflow automation with LLM technology for medical literature processing.While formal validation with quantitative metrics and expert comparison is needed, this work establishes a framework for developing scalable, automated literature review systems.Future studies should evaluate processing efficiency, output quality metrics, and comparison with emerging tools, including free alternatives.